US10568600B2 - System and method for detecting anatomical regions - Google Patents

System and method for detecting anatomical regions Download PDF

Info

Publication number
US10568600B2
US10568600B2 US15/885,088 US201815885088A US10568600B2 US 10568600 B2 US10568600 B2 US 10568600B2 US 201815885088 A US201815885088 A US 201815885088A US 10568600 B2 US10568600 B2 US 10568600B2
Authority
US
United States
Prior art keywords
landmark
landmarks
atlas
result
water equivalent
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US15/885,088
Other languages
English (en)
Other versions
US20190231294A1 (en
Inventor
Pierre Guntzer
Philippe Roy
Nicolas Grussenmeyer
Mathieu Bedez
Yacine El Farouk
Gaetan Fritz
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
General Electric Co
Original Assignee
General Electric Co
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by General Electric Co filed Critical General Electric Co
Priority to US15/885,088 priority Critical patent/US10568600B2/en
Assigned to GENERAL ELECTRIC COMPANY reassignment GENERAL ELECTRIC COMPANY ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: FRITZ, GAETAN, BEDEZ, MATHIEU, ROY, PHILIPPE, GUNTZER, Pierre, EL FAROUK, YACINE, GRUSSENMEYER, NICOLAS
Priority to EP19746607.1A priority patent/EP3745950B1/en
Priority to PCT/US2019/015662 priority patent/WO2019152412A1/en
Priority to CN201980015398.7A priority patent/CN111757701B/zh
Publication of US20190231294A1 publication Critical patent/US20190231294A1/en
Application granted granted Critical
Publication of US10568600B2 publication Critical patent/US10568600B2/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5294—Devices using data or image processing specially adapted for radiation diagnosis involving using additional data, e.g. patient information, image labeling, acquisition parameters
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/48—Diagnostic techniques
    • A61B6/488—Diagnostic techniques involving pre-scan acquisition
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
    • A61B6/5217—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data extracting a diagnostic or physiological parameter from medical diagnostic data
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
    • A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
    • A61B6/5229—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data combining image data of a patient, e.g. combining a functional image with an anatomical image
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/54—Control of apparatus or devices for radiation diagnosis
    • A61B6/545—Control of apparatus or devices for radiation diagnosis involving automatic set-up of acquisition parameters
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/58—Testing, adjusting or calibrating thereof
    • A61B6/582—Calibration
    • A61B6/583—Calibration using calibration phantoms
    • A61B6/584—Calibration using calibration phantoms determining position of components of the apparatus or device using images of the phantom
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • G06T7/0014—Biomedical image inspection using an image reference approach
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/174—Segmentation; Edge detection involving the use of two or more images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G06T7/74—Determining position or orientation of objects or cameras using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/30—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for calculating health indices; for individual health risk assessment
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
    • A61B6/03—Computed tomography [CT]
    • A61B6/032—Transmission computed tomography [CT]
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/48—Diagnostic techniques
    • A61B6/486—Diagnostic techniques involving generating temporal series of image data
    • A61B6/487—Diagnostic techniques involving generating temporal series of image data involving fluoroscopy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
    • A61B6/56—Details of data transmission or power supply, e.g. use of slip rings
    • A61B6/563—Details of data transmission or power supply, e.g. use of slip rings involving image data transmission via a network
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10081—Computed x-ray tomography [CT]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10072—Tomographic images
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10116—X-ray image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20112—Image segmentation details
    • G06T2207/20128—Atlas-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30204—Marker

Definitions

  • This disclosure relates to systems and methods for automatically identifying anatomical regions (e.g., head, shoulders, chest, abdomen, pelvis, etc.) in medical images.
  • anatomical regions e.g., head, shoulders, chest, abdomen, pelvis, etc.
  • ionizing radiation to acquire images of patients and/or provide treatment to patients, such as x-ray devices, computed tomography (CT) devices, fluoroscopic devices, and so on.
  • CT computed tomography
  • a beam of ionizing radiation e.g., x-rays
  • an exposed subject e.g., patient
  • the ionizing radiation is attenuated when passing through the exposed subject and then received at a detector.
  • the attenuation of the ionizing radiation is measured and processed to generate an image that can be viewed and analyzed.
  • Radiation doses can vary significantly between different types of studies, for example, a radiation dose for a routine head CT may be much lower than a dose for a multiphase abdomen and pelvis CT.
  • low dose scout images are acquired first for anatomy localization.
  • a clinician may review the scout images, manually annotate anatomical regions, and set the radiation dose (and other scanning parameters) to acquire images of an anatomical region of interest.
  • anatomical regions can be identified automatically in the scout images without manual intervention and the local dose of radiation can be determined automatically based on some dose optimization strategies. Thus, automatic identification of anatomical regions with improved accuracy is generally desired.
  • the present disclosure provides a method for automatically identifying anatomical regions in medical images.
  • the method comprises generating a signature of an exposed subject.
  • the signature comprises a water equivalent diameter distribution generated from one or more images of the exposed subject.
  • the method also comprises identifying a best matching atlas element from an atlas.
  • the atlas includes a group of atlas elements.
  • Each atlas element includes landmarks associated with a set of image data and a signature associated the set of image data.
  • the signature of the best matching atlas element matches the signature of the exposed subject the best among the atlas.
  • the method further comprises projecting landmarks of the best matching atlas element onto an image of the exposed subject to generate projected landmarks for the image of the exposed subject.
  • the present disclosure provides a processing system for automatically identifying anatomical regions in medical images.
  • the processing system comprises a signature generator configured to generate a signature of an exposed subject.
  • the signature comprises a water equivalent diameter distribution generated from one or more images of the exposed subject.
  • the processing system also comprises a signature matcher configured to identify a best matching atlas element from an atlas.
  • the atlas includes a group of atlas elements.
  • Each atlas element includes landmarks associated with a set of image data and a signature associated the set of image data.
  • the signature of the best matching atlas element matches the signature of the exposed subject the best among the atlas.
  • the processing system further comprises a landmark projector configured to project landmarks of the best matching atlas element onto an image of the exposed subject to generate projected landmarks for the image of the exposed subject.
  • the present disclosure provides a non-transitory computer readable medium comprising instructions which, when executed by a processing system, cause the processing system to perform operations.
  • the operations comprise generating a signature of an exposed subject.
  • the signature comprises a water equivalent diameter distribution generated from one or more images of the exposed subject.
  • the operations also comprise identifying a best matching atlas element from an atlas.
  • the atlas includes a group of atlas elements.
  • Each atlas element includes landmarks associated with a set of image data and a signature associated the set of image data.
  • the signature of the best matching atlas element matches the signature of the exposed subject the best among the atlas.
  • the operations further comprise projecting landmarks of the best matching atlas element onto an image of the exposed subject to generate projected landmarks for the image of the exposed subject.
  • FIG. 1 is a block diagram of the environment for automatically identifying anatomical regions in medical images, in accordance with an exemplary embodiment
  • FIG. 2 is a schematic diagram illustrating the axes with respect to an exposed subject on a table, in accordance with an exemplary embodiment
  • FIG. 3 shows a slice image with respect to scout images, in accordance with an exemplary embodiment
  • FIG. 4 is a block diagram of a first anatomical region detector which can be used in FIG. 1 , in accordance with an exemplary embodiment
  • FIG. 5 illustrates calculating water equivalent diameter distribution by using the slice image and the scout image, in accordance with an exemplary embodiment
  • FIG. 6 is a structure of an atlas element, in accordance with an exemplary embodiment
  • FIG. 7 illustrates landmarks on an image of a human body, in accordance with an exemplary embodiment
  • FIG. 8 is a flow chart of a method for identifying anatomical regions, in accordance with an exemplary embodiment.
  • the articles “a,” “an,” and “the” are intended to mean that there are one or more of the elements.
  • the terms “comprising,” “including,” and “having” are intended to be inclusive and mean that there may be additional elements other than the listed elements.
  • one object e.g., a material, element, structure, member, etc.
  • references to “one embodiment” or “an embodiment” of the present disclosure are not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the recited features.
  • the present disclosure is to provide systems and methods for automatically identifying anatomical regions in medical images.
  • the systems and methods disclosed herein are used in conjunction with another anatomy localization system/method to improve the accuracy of anatomy identification, for example, to remove false positives from results of the another method.
  • the disclosure is used on its own for identifying anatomical regions.
  • An exemplary method includes generating a signature for a patient.
  • the signature includes patient specific metrics and does not depend on the model and/or protocol used to acquire the images.
  • the signature includes a distribution of water equivalent diameter generated from one or more images of the patient.
  • the signature may also include other patient specific metrics, such as sex, age, size, etc.
  • the signature of the patient is compared to the same kind of signatures of a set of atlas elements (or phantoms).
  • An atlas includes this set of elements, each having image data and associated signature. Positions of the boarders of anatomical stages (also known as landmarks) have already been annotated for each atlas element.
  • An atlas element for which the signature matches the signature of the patient the best is identified from the atlas.
  • the landmarks of the identified best matching atlas element are projected onto the patient image. If the method disclosed herein is used in conjunction with another anatomy localization method, the projected landmarks on the patient image can be compared to the landmarks determined by the another method. A confidence score is given based on the comparison to each landmark identified by the another method to indicate how likely the landmark is correct.
  • the confidence score associated with this landmark indicates “unlikely.”
  • the “unlikely” landmark can be removed from the results of the existing method.
  • the “unlikely” landmark can be replaced by the corresponding projected landmark.
  • the projected landmark can be used to complement the result.
  • the disclosure does not require that the image of the patient at issue and the atlas images are produced by the same model/protocol.
  • two images produced by different models/protocols are difficult to compare because the gray levels of an image highly depend upon the machine/protocol.
  • the patient specific signature being used for comparison, a wide set of atlas element coming from various devices can be used as atlas elements. That is to say, the disclosure works for images generated by an imaging device for which no atlas element is present in the atlas because device manufacturer or model does not affect the patient specific metrics in the signature.
  • the environment 100 comprises an imaging device 110 , a picture archiving and communication system (PACS) 120 , a first anatomical region detector 130 , and optionally a second anatomical region detector 140 , connected via communication connection 150 .
  • the imaging device 110 may employ a beam of ionizing radiation 112 to acquire medical images for an exposed subject 114 resting on a table 116 .
  • the medical images may be stored in the PACS 120 .
  • the first anatomical region detector 130 may identify anatomical regions in the medical images with reference to an atlas 132 .
  • the first anatomical region detector 130 works in conjunction with the second anatomical region detector 140 to correct the results of the second anatomical region detector 140 .
  • the imaging device 110 can be an x-ray (e.g., angiographic) imaging device, CT device, fluoroscopic imaging device, or any other medical imaging device that use ionizing radiation (e.g., x-rays) to acquire images.
  • a beam of ionizing radiation 112 is projected towards the exposed subject 114 (e.g., patient) resting on the table 116 .
  • the ionizing radiation 112 is attenuated when passing through the exposed subject 114 and then received at a detector or a photographic plate.
  • the attenuation of the ionizing radiation 112 is measured and processed to generate an image that can be viewed and analyzed.
  • the imaging device 110 first uses low dose of radiation to perform an initial topographic scan. Slice images and one or more scout images on the exposed subject 114 can be acquired.
  • a scout image is a two-dimensional image, which is a general illustration of a projected view of the exposed subject 114 .
  • two scout images of the same scene are acquired under different points of view, such as a lateral (LAT) view and an anteroposterior (AP) view, which allows reconstruction of a three-dimensional outer shape of exposed subject 114 .
  • FIG. 2 shows an exemplary coordinate system that the imaging device 110 uses, to explain the LAT view and the AP view.
  • X-axis is a horizontal axis extending in a direction from a left-hand border to a right-hand border of a surface of the table 116
  • y-axis is a vertical axis extending in a direction from the floor upwards in direction towards a top of the table 116
  • z-axis is a longitudinal axis extending in a direction from head to toe of the exposed subject 114 .
  • the LAT view is a projected view of the exposed subject 114 in the YZ plane
  • the AP view is a projected view of the exposed subject 114 in the XZ plane.
  • FIG. 3 shows an exemplary slice image with respect to scout images.
  • a slice image 330 shows an elliptical representation 332 of the exposed subject 114 as a cross-sectional image.
  • the elliptical representation 332 corresponds to a graphic line 312 in the LAT scout image 310 and a graphic line 322 in the AP scout image 320 .
  • a user can select a region of interest from the scout images 310 and 320 by changing the positions of the graphic lines 312 and 314 and view the corresponding slice images.
  • the imaging device 110 generates and/or stores the scout images in the Digital Imaging and Communications in Medicine (DICOM) format.
  • DICOM Digital Imaging and Communications in Medicine
  • Other formats can also be used, such as JPEG, TIFF, GIF, PNG, etc.
  • the scout image data includes a collection or series of tags where each tag includes pixel data having a value that represents the pixel (e.g., basic unit of programmable color on a display monitor) associated with the tag.
  • the scout image data may also include metadata associated with patient specific metrics, such as sex, age, size, and so on.
  • the PACS 120 stores medical images (e.g., scout images, slice images) acquired by the imaging device 110 in a database or registry.
  • the images are stored in PACS 120 by healthcare practitioners (e.g., technicians, physicians, radiologists) after a medical imaging of a patient and/or automatically transmitted from imaging device 110 to PACS 120 for storage.
  • the first anatomical region detector 130 can receive images of the exposed subject 114 from the imaging device 110 and/or the PACS 120 , determine a signature of the exposed subject from the images, identify an atlas element from the atlas 132 for which the signature matches the signature of the exposed subject 114 the best, and project landmarks of the best matching atlas element onto an image of the exposed subject.
  • the first anatomical region detector 130 works in conjunction with the second anatomical region detector 140 to correct the results of the second anatomical region detector 140 . Structures of the first anatomical region detector 130 and the atlas element will be explained in detail below with reference to FIGS. 4 and 6 .
  • the second anatomical region detector 140 can be any appropriate detector that uses a different method than the first detector 130 to identify anatomical regions in the scout image(s) of the exposed subject 114 received from the imaging device 110 and/or the PACS 120 .
  • the second anatomical region detector 140 can rely on the first anatomical region detector 130 to correct its results of region detection.
  • An exemplary method for identifying anatomical regions that can be used by the second detector 140 is described in U.S. patent application Ser. No. 14/510,815 (issued as U.S. Pat. No. 9,649,079, “the '079 patent”), which is incorporated herein in its entirety by reference.
  • the method described in the patent uses an AP scout image of an exposed subject acquired by a CT device.
  • a contour of an extremity of the exposed subject is calculated from the AP scout image, and the contour is delineated by anatomical landmarks (i.e., boarders of the anatomical stages) that mark an abrupt change in the thickness of the exposed subject.
  • Other landmarks that might not correspond to an abrupt change in the thickness can be identified by applying heuristic and/or auxology criteria to recognize some specific shapes in the exposed subject.
  • Results of the method disclosed in the '079 patent may sometimes include false positives because in general, the method tries to identify a pattern in the AP scout image and returns a portion of the image that corresponds to this pattern the best. However, if the pattern is not present in the image, the method may still return a false positive.
  • the scout image may include a very small part of the pelvic region or may not contain the pelvic region but other shapes that look like the missing pelvis. Errors can occur in two situations. First, the method may use the maximum of an energy function to finds out the “best candidate” for the start of the pelvic region. If the pelvis is not included in the scout image, this maximum is irrelevant. Second, the method may identify a region that looks like a pelvis but is not a pelvis. In these situation, the second detector 140 can rely on the first detector 130 to remove the false positives, as will be discussed below in detail.
  • the communication connection 150 facilitates communication among the devices and systems shown in FIG. 1 .
  • the communication connection 150 can be implemented by any appropriate wired connection, wireless connection, or any combination therefore, such as cable, data bus, universal serial bus (USB) connection, radio frequency (RF) connection, infrared connection, optical connection, near field communication (NFC), wide area network (WAN), local area network (LAN), the Internet, a cloud-based computing infrastructure, etc.
  • USB universal serial bus
  • RF radio frequency
  • NFC near field communication
  • WAN wide area network
  • LAN local area network
  • the Internet a cloud-based computing infrastructure, etc.
  • the environment 100 for identifying anatomical regions as shown in FIG. 1 is for illustration, not for limitation. Any appropriate environment can include more, fewer, and/or different components. Some components can be integrated, for example, the first anatomical region detector 130 can be integrated with the second anatomical region detector 140 . Variations are contemplated by this disclosure.
  • the first anatomical region detector 400 may be used as the first anatomical region detector 130 in FIG. 1 .
  • the first anatomical region detector 400 may be implemented on any appropriate processing system, such as system-on-a-chip system, personal computer, workstation, cloud-based computing system, and so on.
  • the first anatomical region detector 400 may be implemented on more than one processing system, i.e., a portion of the first detector 400 is run on one processing system, and another portion of the first detector 400 is run on another processing system.
  • the first anatomical region detector 400 comprises a signature generator 402 , signature matcher 404 , landmark projector 406 , and optionally a result comparator 408 , and result corrector 410 .
  • the signature generator 402 generates a signature of an exposed subject (e.g., patient) from one or more images of the exposed subject.
  • the signature matcher 404 identifies an atlas element from the atlas for which the signature matches the signature of the exposed subject the best.
  • the landmark projector 406 projects landmarks of the best matching atlas element onto an image of the exposed subject.
  • the result comparator 408 compares the projected landmarks to the result of the second anatomical region detector 140 .
  • the result corrector 410 corrects the result of the second detector 140 based on the comparison.
  • the signature generator 402 generates a signature that includes a water equivalent diameter distribution from one or more images of the exposed subject.
  • the x-ray attenuation of the exposed subject can be expressed in terms of a water cylinder having the same x-ray absorption.
  • the area and diameter of such a cylinder of water are referred to as the water equivalent area (A w ) and water equivalent diameter (D w ).
  • the water equivalent diameter distribution refers to the various water equivalent diameters D w along the longitudinal axis (i.e., the z-axis in FIG. 2 ).
  • Various methods may be used to calculate the water equivalent diameters D w along the longitudinal axis.
  • an article titled “Use of Water Equivalent Diameter for Calculating Patient Size and Size-Specific Dose Estimates (SSDE) in CT”, the Report of AAPM Task Group 220 discloses a method for calculating water equivalent diameter from CT slice images.
  • the disclosure herein uses a new method for calculating water equivalent diameter distribution which is faster than the method disclosed in the article.
  • FIG. 5 illustrates the method for fast calculating water equivalent diameter distribution, in accordance with an exemplary embodiment.
  • the method uses a few (e.g., 3, 4, 5, etc.,) slice images 530 and one or more scout images 520 .
  • the slice images are expressed in CT numbers (i.e., normalized to the linear attenuation of water) with the unit of Hounsfield Units (HU).
  • CT numbers i.e., normalized to the linear attenuation of water
  • HU Hounsfield Units
  • CT(x,y) ROI is the mean CT number in the region of interest (ROI)
  • a ROI is the area of the ROI.
  • the pixel values of a scout image are not normalized to absolute units, but in grey levels, for example.
  • Each pixel (e.g., A, B) in the AP scout image 520 is associated with an x-ray path from the source 502 , through the exposed subject, to the detector.
  • the pixel value is proportional to the total attenuation along this x-ray path.
  • An estimate of the overall water equivalent diameter at the position of line AB is proportional to the sum of the pixel values within the bound of the exposed subject.
  • the relationship between the water equivalent diameter and pixel values can vary among device models/protocols.
  • a slice image is a section along the z-axis which corresponds to a line of a scout image
  • the relationship between the grey levels of a particular scout image and the water equivalent diameter can be calibrated by the water equivalent diameters calculated from the slice images.
  • the scout image 530 shows the location (i.e., line AB) of the slice image 520 with respect to the exposed subject 114 .
  • the water equivalent diameter D w of the slice image 530 can be calculated according to equation (1).
  • the sum of grey levels along line AB corresponds to the calculated D w .
  • water equivalent diameters calculated from a few (e.g., 3, 4, 5, etc.) slice images and grey level sums at corresponding positions of the scout image can be fit into a linear relationship.
  • the sum of gray levels within the boundary can be calculated and translated into water equivalent diameter using, for example, an interpolation.
  • the water equivalent diameter distribution can be expressed as a curve.
  • a curve 542 in a graph 540 shows the water equivalent diameter distribution along the z-axis.
  • the distribution can also be expressed in other forms, e.g., lookup table.
  • FIG. 5 shows that an AP scout image 520 is used to calculate the water equivalent diameter distribution, it should be understood that a different scout image (e.g., LAT scout) can be used for the calculation. In some embodiments, more than one scout images can be used.
  • the sum of attenuations may include contributions from voxels of the exposed subject and the surrounding media (e.g., table, air, clothes, mattress, etc.).
  • the contribution of the exposed subject can be isolated from the surrounding media when the water equivalent diameter is calculated.
  • the signature generator 402 may apply convolution filters (e.g., median filters, Gaussian blur filters, etc.) to the scout image data before calculating the water equivalent diameter distribution. Pre-processing the scout image can aid in determining the boundaries of the exposed subject. For example, the signature generator 402 may calculate the boundaries by comparing or analyzing a greyscale intensity of candidate pixels.
  • convolution filters e.g., median filters, Gaussian blur filters, etc.
  • the signature may include other patient specific metrics besides the water equivalent diameter distribution, such as age, sex, size, etc.
  • the signature may be a weighted combination of these metrics.
  • the signature matcher 404 identifies a best matching atlas element from the atlas 132 by comparing the signature of the exposed subject matter to the signature of each atlas element.
  • the atlas 132 includes a set of images of the same kind (e.g., AP or LAT scout) and each particular image in the atlas is called an atlas element.
  • FIG. 6 shows an exemplary structure of an atlas element.
  • an atlas element includes a set of image data, signature associated with the set of image data, and landmarks associated with the set of image data.
  • an atlas element does not include the image data, but includes the signature and the landmarks.
  • the signature may include the water equivalent diameter distribution (e.g., curve, lookup table) calculated for the image data, and other metadata such as age, sex, size, etc. of the subject of the image data.
  • the landmarks annotate the positions of borders of anatomical stages (e.g., head, shoulder, chest, abdomen, pelvis, lower limb, etc.) for the image data. In some embodiments, the landmarks were annotated or corrected manually by a clinician beforehand.
  • the atlas 132 comprises a pre-selected set of various patients.
  • the atlas 132 may include various numbers of atlas elements (e.g., 20, 30, 50, 100, etc.).
  • the imaging device 110 used to acquire the image(s) of the exposed subject 110 may not have an atlas element present in the atlas 132 . In other words, because the signature contains only patient specific metrics, atlas element from different manufacturers or models can be used.
  • WED j atlas element
  • WED j patient
  • WED j patient
  • WED j patient
  • This equation represents the line-by-line sum of the difference between the water equivalent diameter of the patient and the water equivalent diameter of the atlas element.
  • the score is computed based on partial matching between an atlas element and the patient signature. For example, the area from which the patient signature is generated might correspond to only 80% of the area of the atlas element, or vice versa. In these situations, the score may be computed for the overlapping areas.
  • the signature includes multiple metrics
  • S i is a metric in the signature (i.e., water equivalent diameter distribution, age, sex, size, etc.), and Weight i is a predefined weight given to the metric.
  • This equation represents the weighted sum of the difference of each signature component between the atlas element and the exposed subject 114 .
  • the atlas element with the lowest score is determined to be the best match for the exposed subject 114 at issue.
  • a threshold score is set. If the lowest score of the atlas 132 is greater than the threshold score, the signature matcher 404 may determine that there is not an atlas element stored in the atlas 132 that matches or nearly matches the image of the exposed subject 114 . In further embodiments, the signature matcher 404 may use a different atlas comprising a different set of atlas elements for the matching in this situation. In some embodiments, the atlas 132 may be enhanced to include more atlas elements.
  • the landmark projector 406 projects the landmarks of the identified besting matching atlas element onto the image of the exposed subject 114 to generate projected landmarks on the image. As discussed above, landmarks have been annotated for the atlas element.
  • the landmark projector 406 may extract various landmarks (i.e., positions) from best matching atlas element, identify corresponding positions on the image of the exposed subject (e.g., positions of corresponding z-value), and annotate the corresponding positions.
  • FIG. 7 shows the landmarks and anatomical stages on an AP scout image of a human body, in accordance with an exemplary embodiment.
  • the contours 702 and 704 define the boundaries of the exposed subject.
  • Line 712 marks the start of chest 722
  • line 714 marks the end of chest 722 and the start of abdomen 724
  • line 716 marks the end of abdomen 724 and the start of pelvis 726
  • line 718 marks the end of pelvis 726 and the start of lower limb 728 .
  • Lines 712 , 714 , 716 , and 718 are landmarks that mark the positions of borders of anatomical stages 722 , 724 , 726 , and 728 . It should be understood that FIG. 7 is for illustration not for limitation. An image may not include all the landmarks but only one or some of the landmarks.
  • the result of the first anatomical region detector 400 is used on its own for automatic anatomy localization. In some embodiments, the result of the first detector 400 is used to correct the result of the second detector 140 .
  • the first detector 400 may further comprises a result comparator 408 . In some embodiments, the result comparator 408 compares the landmarks identified by the second detector 140 to the corresponding landmarks identified by the first detector 400 and gives a confidence score to each landmark based on the comparison. The confidence score indicates how likely the landmark identified by the second detector 140 is correct.
  • the result comparator 408 determines the difference in z-axis values between a landmark (e.g., landmark 712 , 714 , 716 , or 718 ) determined by the landmark projector 406 and a corresponding landmark determined by the second detector 140 . If the difference is substantial (i.e., the two landmarks are far away from each other), the result comparator 408 gives a low confidence score to indicate that the landmark determined by the second detector 140 is “unlikely” to be correct. If the difference is small (i.e., the two landmarks are close), the result comparator 408 give a high confidence score to indicate that the landmark determined by the second detector 140 is “likely” to be correct.
  • a landmark e.g., landmark 712 , 714 , 716 , or 718
  • the confidence score has a negative association with the difference, i.e., the greater the difference, the lower the confidence score.
  • the result comparator 408 compares the difference with a predetermined threshold. If the difference is greater than the threshold, the confidence determiner 408 gives a first confidence score. If the difference is smaller than the threshold, the confidence determiner 408 give a second confidence score, which is higher than the first confidence score. In some embodiments, if a landmark identified by the second detector 140 does not find a counterpart in the result of the landmark projector 406 , the confidence determiner 408 gives a low confidence score to indicate so.
  • the first detector 400 may further comprise a result corrector 410 configured to correct the results of the second detector 140 .
  • the result corrector 410 may remove or replace that particular landmark.
  • the result corrector 410 compares each confidence score against a predefined confidence threshold. If the confidence score is lower than the predefined confidence threshold, the result corrector 410 corrects the associated landmark. For example, if the second detector 140 identifies a landmark which does not have a counterpart in the results of the first detector 400 , the result corrector 410 may remove this landmark from the results of the second detector 140 .
  • the result corrector 410 may replace the landmark determined by the second detector 140 with the corresponding landmark determined by the first detector 400 .
  • the second detector 140 might have failed to identify a landmark that is identified by the first detector 400 .
  • the first detector 400 identifies a border between the pelvis and the lower limb (e.g., landmark 728 in FIG. 7 ) while the second detector 140 failed to detect such a border.
  • the result corrector 410 may add this landmark to the results of the second detector 410 .
  • an anatomical region in the best matching atlas element corresponds roughly to the same anatomical region in the real patient, which allows modification of the energy function used or deduction of the anatomical region not included in the patient image. The closer the atlas element matches the patient image, the more accurate correction can be achieved.
  • a flow chart 800 of a method for identifying anatomical regions is shown, in accordance with an exemplary embodiment.
  • the method can be executed by the first anatomical region detector 130 in FIG. 1 and the first anatomical region detector 400 in FIG. 4 .
  • a signature is computed for an exposed subject from one or more images of the exposed subject.
  • the signature includes patient specific metrics and does not rely upon the machine and/or protocol to acquire the image(s).
  • the signature includes a water equivalent diameter distribution calculated from the image(s) using the fast method as discussed above.
  • the water equivalent diameter distribution is generated using another appropriate method. For example, certain devices may give the information as metadata in the scout image(s).
  • the signature may further include age, sex, size, etc., of the exposed subject.
  • a best matching atlas element is identified from an atlas, where a signature of the best matching atlas element matches the signature of the exposed subject the best.
  • the atlas includes a group of images, each called an atlas element.
  • Each atlas element includes image data, associated signature, and associated landmarks that have been annotated.
  • the difference between the signature of the exposed subject and the signature of each atlas element is calculated.
  • the atlas element with the minimum difference is identified to be the best matching atlas element.
  • landmarks of the best matching atlas element are projected onto an image of the exposed subject to create projected landmarks for the image.
  • the z-axis values of the landmarks are mapped onto corresponding z-axis positions of the image.
  • the projected landmarks can be used alone for anatomy localization—and operations 808 through 820 may not be needed.
  • the projected landmarks are used to correct the result of another anatomy localization method and at least some of operations 808 through 820 can be performed.
  • the projected landmarks are compared to a second set of landmarks identified by the another method.
  • difference in z-axis values between a projected landmark and a corresponding landmark in the second set of landmarks is calculated.
  • a confidence score is determined based on the difference. In some embodiments, the greater the difference, the lower the confidence score.
  • a first confidence score is given to the difference greater than a predefined threshold, and a second confidence score is given to the difference not greater than the predefined threshold, where the first confidence score is lower than the second confidence score.
  • a special confidence score is given to indicate so.
  • any landmark identified by the second method is “unlikely” to be correct based on the comparison.
  • a confidence score associated with each landmark of the second method is compared with a predefined threshold score. If the confidence score is lower than the threshold score, the associated landmark identified by the second method is determined “unlikely” to be correct. If the confidence score is not lower than the threshold score, the associated landmark identified by the second method is not labeled as “unlikely.” If it is determined at operation 810 that no landmark in the result of the second method is “unlikely,” the method proceeds to an operation 818 , which will be discussed later below.
  • a particular landmark in the second set of landmarks is “unlikely,” then at an operation 812 , it is determined whether that particular landmark has a corresponding projected landmark.
  • the another method may have identified a landmark that marks the end of the pelvis and the start of the lower limb while the projected landmarks do not include a corresponding landmark. If this is the case, the particular landmark is removed from the second set of landmarks, at an operation 814 .
  • the “unlikely” landmark is replaced by the corresponding one in the projected landmarks.
  • the second method has identified a landmark that marks the end of the pelvis and the start of the lower limb.
  • the z-axis value of the landmark is substantially different from the z-axis value of the corresponding projected landmark. This particular landmark in second set of landmarks can be replaced by the corresponding projected landmark, at an operation 816 .
  • the method 800 may further include an operation 818 , where it is determined whether all projected landmarks are in the second set of landmarks identified by the another anatomy localization method.
  • the projected landmarks may include a landmark that mark the end of the pelvis and the start of the lower limb while the second set of landmarks do not include a counterpart.
  • the missing landmark can be added to the second set of landmarks, at an operation 820 . If it is determined at operation 818 that all projected landmarks have a corresponding one in the second set of landmarks, then at operation 822 , no further change is made to the result.
  • FIG. 8 is for illustration not for limitation. An appropriate process may include more, fewer, or different operations than those shown in FIG. 8 .

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Radiology & Medical Imaging (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Public Health (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biomedical Technology (AREA)
  • Pathology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Optics & Photonics (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Veterinary Medicine (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Quality & Reliability (AREA)
  • Physiology (AREA)
  • Epidemiology (AREA)
  • Primary Health Care (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
US15/885,088 2018-01-31 2018-01-31 System and method for detecting anatomical regions Active 2038-04-29 US10568600B2 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
US15/885,088 US10568600B2 (en) 2018-01-31 2018-01-31 System and method for detecting anatomical regions
EP19746607.1A EP3745950B1 (en) 2018-01-31 2019-01-29 System and method for detecting anatomical regions
PCT/US2019/015662 WO2019152412A1 (en) 2018-01-31 2019-01-29 System and method for detecting anatomical regions
CN201980015398.7A CN111757701B (zh) 2018-01-31 2019-01-29 用于检测解剖区域的系统和方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US15/885,088 US10568600B2 (en) 2018-01-31 2018-01-31 System and method for detecting anatomical regions

Publications (2)

Publication Number Publication Date
US20190231294A1 US20190231294A1 (en) 2019-08-01
US10568600B2 true US10568600B2 (en) 2020-02-25

Family

ID=67391740

Family Applications (1)

Application Number Title Priority Date Filing Date
US15/885,088 Active 2038-04-29 US10568600B2 (en) 2018-01-31 2018-01-31 System and method for detecting anatomical regions

Country Status (4)

Country Link
US (1) US10568600B2 (zh)
EP (1) EP3745950B1 (zh)
CN (1) CN111757701B (zh)
WO (1) WO2019152412A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210361249A1 (en) * 2020-05-21 2021-11-25 Tata Consultancy Services Limited Anatomical landmark detection and identification from digital radiography images containing severe skeletal deformations
US11908174B2 (en) 2021-12-30 2024-02-20 GE Precision Healthcare LLC Methods and systems for image selection

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111243026B (zh) * 2020-01-19 2023-09-22 武汉联影智融医疗科技有限公司 解剖标记点定位方法、装置、计算机设备和存储介质
CN112262440A (zh) * 2020-07-15 2021-01-22 北京肿瘤医院(北京大学肿瘤医院) 一种通过影像组学特征判断癌症治疗反应的方法及系统
EP4059435A1 (en) * 2021-03-15 2022-09-21 Koninklijke Philips N.V. Patient preparation for medical imaging

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030007601A1 (en) * 2000-02-18 2003-01-09 Jaffray David A. Cone-beam computerized tomography with a flat-panel imager
US20100303314A1 (en) * 2009-05-29 2010-12-02 Siemens Medical Solutions Usa, Inc. Systems and methods for detecting and visualizing correspondence corridors on two-dimensional and volumetric medical images
US8045771B2 (en) 2006-11-22 2011-10-25 General Electric Company System and method for automated patient anatomy localization
US20120070053A1 (en) * 2010-09-22 2012-03-22 Siemens Aktiengesellschaft Method and System for Vascular Landmark Detection
US20150305695A1 (en) * 2014-04-25 2015-10-29 Medtronic, Inc. Guidance System For Localization And Cannulation Of the Coronary Sinus
US9649079B1 (en) 2014-10-09 2017-05-16 General Electric Company System and method to illustrate a radiation dose applied to different anatomical stages of an exposed subject
US20170143291A1 (en) 2015-11-19 2017-05-25 General Electric Company Water equivalent diameter determination from scout images
US20180350454A1 (en) * 2017-05-30 2018-12-06 Siemens Healthcare Gmbh Determining an anonymized dose report image
US20180368785A1 (en) * 2015-12-17 2018-12-27 Koninklijke Philips N.V. Method for estimating the radiation dose received by an organ during a computed tomography scan

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2013152744A (ru) * 2011-04-28 2015-06-10 Конинклейке Филипс Н.В. Медицинское устройство формирования изображений с отдельной кнопкой для выбора изображений-кандидатов сегментации
IN2014MN02601A (zh) * 2012-06-04 2015-07-24 Tel Hashomer Medical Res Infrastructure & Services Ltd
DE102014211150A1 (de) * 2014-06-11 2015-12-17 Siemens Aktiengesellschaft Verfahren zum Betreiben eines Computertomographen und Computertomograph
US9949643B2 (en) * 2014-10-18 2018-04-24 International Business Machines Corporation Automatic visualization of regional functional parameters of left ventricle from cardiac imaging
CN107497059A (zh) * 2017-07-12 2017-12-22 合肥中科离子医学技术装备有限公司 一种质子ct精准控制方法

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030007601A1 (en) * 2000-02-18 2003-01-09 Jaffray David A. Cone-beam computerized tomography with a flat-panel imager
US8045771B2 (en) 2006-11-22 2011-10-25 General Electric Company System and method for automated patient anatomy localization
US20100303314A1 (en) * 2009-05-29 2010-12-02 Siemens Medical Solutions Usa, Inc. Systems and methods for detecting and visualizing correspondence corridors on two-dimensional and volumetric medical images
US20120070053A1 (en) * 2010-09-22 2012-03-22 Siemens Aktiengesellschaft Method and System for Vascular Landmark Detection
US20150305695A1 (en) * 2014-04-25 2015-10-29 Medtronic, Inc. Guidance System For Localization And Cannulation Of the Coronary Sinus
US9649079B1 (en) 2014-10-09 2017-05-16 General Electric Company System and method to illustrate a radiation dose applied to different anatomical stages of an exposed subject
US20170143291A1 (en) 2015-11-19 2017-05-25 General Electric Company Water equivalent diameter determination from scout images
US20180368785A1 (en) * 2015-12-17 2018-12-27 Koninklijke Philips N.V. Method for estimating the radiation dose received by an organ during a computed tomography scan
US20180350454A1 (en) * 2017-05-30 2018-12-06 Siemens Healthcare Gmbh Determining an anonymized dose report image

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
McCollough et al., "Use of Water Equivalent Diameter for Calculating Patient Size and Size-Specific Dose Estimates (SSDE) in CT," The Report of AAPM Task Group 220, Sep. 2014, 23 pages.

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20210361249A1 (en) * 2020-05-21 2021-11-25 Tata Consultancy Services Limited Anatomical landmark detection and identification from digital radiography images containing severe skeletal deformations
US11696736B2 (en) * 2020-05-21 2023-07-11 Tata Consultancy Services Limited Anatomical landmark detection and identification from digital radiography images containing severe skeletal deformations
US11908174B2 (en) 2021-12-30 2024-02-20 GE Precision Healthcare LLC Methods and systems for image selection

Also Published As

Publication number Publication date
CN111757701A (zh) 2020-10-09
CN111757701B (zh) 2022-03-04
WO2019152412A1 (en) 2019-08-08
EP3745950A1 (en) 2020-12-09
US20190231294A1 (en) 2019-08-01
EP3745950A4 (en) 2021-11-03
EP3745950B1 (en) 2023-03-22

Similar Documents

Publication Publication Date Title
US10568600B2 (en) System and method for detecting anatomical regions
McCollough et al. Use of water equivalent diameter for calculating patient size and size-specific dose estimates (SSDE) in CT: the report of AAPM task group 220
Gan et al. Toward accurate tooth segmentation from computed tomography images using a hybrid level set model
US9918691B2 (en) Device and method for determining image quality of a radiogram image
US10154820B2 (en) Method and apparatus for navigating CT scan with a marker
US9336613B2 (en) Apparatus for generating assignments between image regions of an image and element classes
JP2011508620A (ja) 特徴に基づいた２次元／３次元画像のレジストレーション
JP6313024B2 (ja) スカウト画像内でローカライザを自動的に決定するための方法およびシステム
US20080123929A1 (en) Apparatus, method and program for image type judgment
US11779288B2 (en) Methods, systems, and apparatus for determining radiation doses
CN111260647B (zh) 一种基于图像检测的ct扫描辅助方法、计算机可读存储介质、ct扫描装置
Yin et al. Acquisition, preprocessing, and reconstruction of ultralow dose volumetric CT scout for organ‐based CT scan planning
Salimi et al. Fully automated accurate patient positioning in computed tomography using anterior–posterior localizer images and a deep neural network: a dual-center study
Leyden et al. Cross-correlation template matching for liver localisation in computed tomography
Gottschalk et al. Deep learning based metal inpainting in the projection domain: Initial results
EP4125031A1 (en) Method and systems for removing anti-scatter grid artifacts in x-ray imaging
US20240029256A1 (en) Determining target object type and position
US11948389B2 (en) Systems and methods for automatic detection of anatomical sites from tomographic images
EP4287201A1 (en) Compensating for differences in medical images
JP2010017370A (ja) 濃度調整装置、濃度調整方法、及びプログラム
Lee et al. Estimation of anatomical locations using standard frame of reference in chest CT scans
Brehler Intra-operative visualization and assessment of articular surfaces in C-arm computed tomography images

Legal Events

Date Code Title Description
FEPP Fee payment procedure

Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

AS Assignment

Owner name: GENERAL ELECTRIC COMPANY, NEW YORK

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GUNTZER, PIERRE;ROY, PHILIPPE;GRUSSENMEYER, NICOLAS;AND OTHERS;SIGNING DATES FROM 20180126 TO 20180130;REEL/FRAME:045144/0978

STPP Information on status: patent application and granting procedure in general

Free format text: NON FINAL ACTION MAILED

STPP Information on status: patent application and granting procedure in general

Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER

STPP Information on status: patent application and granting procedure in general

Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS

STPP Information on status: patent application and granting procedure in general

Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED

STCF Information on status: patent grant

Free format text: PATENTED CASE

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 4